when does a central bank’s balance sheet require …/media/documents/... · abstract.the large...
TRANSCRIPT
WHEN DOES A CENTRAL BANK’S BALANCE SHEET REQUIREFISCAL SUPPORT?
MARCO DEL NEGRO AND CHRISTOPHER A. SIMS
ABSTRACT. The large balance sheet of many central banks has raised con-
cerns that they could suffer significant losses if interest rates rose, and
might need a capital injection from the fiscal authority. This paper con-
structs a simple deterministic general equilibrium model, calibrated to US
data, to study the the impact of alternative interest rates scenarios on the
central bank’s balance sheet. We show that the central bank’s policy rule
(or, equivalently, inflation objectives), and the behavior of seigniorage un-
der high inflation are crucial in determining whether a capital injection
might be needed. We show also that a central bank that is seen as ready, in
order to avoid a need for capital injection, to create seignorage by altering
its inflation objectives, can thereby lose control of the price level.
We conclude that for current balance sheet levels of the Federal Reserve
system, direct treasury support would be necessary only under what seem
rather extreme scenarios. However this result depends on assumptions
about demand for non-interest bearing Fed liabilities (mainly currency)
that cannot be firmly grounded in empirical estimates. Higher balance
sheet levels, or a lower currency demand than assumed here could force
the central bank to request a capital injection in order to maintain its infla-
tion control policy.
Date: April 14, 2014; First Draft: April 2013.
PRELIMINARY AND INCOMPLETE DRAFT
Marco Del Negro, Federal Reserve Bank of New York, [email protected].
Christopher A. Sims, Princeton University, [email protected]
We thank Seth Carpenter, and seminar participants at the Bank of Canada, ECB, and FRB
Philadelphia for very helpful comments.The views expressed in this paper do not neces-
sarily reflect those of the Federal Reserve Bank of New York or the Federal Reserve System.1
JEL CLASSIFICATION: E58, E59
KEY WORDS: central bank’s balance sheet, solvency
I. INTRODUCTION
Hall and Reis (2013) and Carpenter, Ihrig, Klee, Quinn, and Boote (2013)
have explored the likely path of the Federal Reserve System’s balance sheet
during a possible return to historically normal levels of interest rates. Both
conclude that, though a period when the system’s net worth at market value
is negative might occur, this is unlikely, would be temporary and would not
create serious problems.1 Those conclusions rely on extrapolating into the
future not only a notion of historically normal interest rates, but also of
historically normal relationships between interest rates, inflation rates, and
components of the System’s balance sheet. In this paper we look at com-
plete, though simplified, economic models in order to study why a central
bank’s balance sheet matters at all and the consequences of a lack of fiscal
backing for the central bank. These issues are important because they lead
us to think about unlikely, but nonetheless possible, sequences of events
that could undermine economic stability. As recent events should have
taught us, historically abnormal events do occur in financial markets, and
understanding in advance how they can arise and how to avert or mitigate
them is worthwhile.2
Constructing a model that allows us to address these issues requires us
to specify monetary and fiscal policy behavior and to consider how demand
1Christensen, Lopez, and Rudebusch (2013) study the interest rate risk faced by the Fed-
eral Reserve using probabilities for alternative interest rate scenarios obtained from a dy-
namic term structure model. They reach the similar conclusions as Hall and Reis (2013)
and Carpenter, Ihrig, Klee, Quinn, and Boote (2013).2A number of recent papers, including Corsetti and Dedola (2012) and Bassetto and
Messer (2013), also study the central bank’s and the fiscal authorities’ balance sheets
separately.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 1
for non-interest-bearing liabilities of the central bank (like currency, or re-
quired reserves paying zero interest) responds to interest rates. Results are
sensitive to these aspects of the specification, even when we consider a ver-
sion of the model calibrated to the current situation in the US.
In the first section below we consider a stripped down model to show
how the need for fiscal backing arises. In subsequent sections we make the
model more realistic and calibrate it to allow simulation of the US Federal
Reserve System’s response to realistic shifts in the real rate or “inflation
scares”.
In both the simple model and the more realistic one, we make some of
the same generic points.
• No policy undertaken by a central bank alone, without fiscal powers,
can guarantee a uniquely determined price level. Cochrane (2011)
has made this point carefully.
• There are simple, plausible, “backstop” fiscal and monetary policy
actions that will make the price level determinate. A commitment
to such actions eliminates the need for overt fiscal backing on the
economy’s equilibrium path.3
• When the price level is uniquely determined, it is nonetheless possi-
ble that the central bank, if its balance sheet is sufficiently impaired,
may need recapitalization in order to maintain its commitment to a
policy rule or an inflation target.
• It is important to distinguish the fiscal “backing” needed to fix the
price level from the fiscal “support” that a separate central bank may
need in order to implement its policy.
3Cochrane has argued that such actions are either impossible, or necessarily so drastic
as to be implausible. One contribution of this paper is to provide a counterexample to
Cochrane’s claim.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 2
• A central bank’s ability to earn seigniorage can make it possible for
it to recover from a situation of negative net worth at market value
without recapitalization from the treasury, while still maintaining its
policy rule. Whether it can do so depends on the policy rule, the de-
mand for its non-interest-bearing liabilities, and the size of the initial
net worth gap.
In the simple model we aim to explain qualitatively how the need for back-
ing or support can arise, while in the more realistic model we try to deter-
mine how likely it is that the US Federal Reserve System will need fiscal
support if interest rates return to more historically normal levels in the near
future.
II. THE SIMPLE MODEL
A central bank that issues fiat money requires fiscal backing if it is to
control the price level. To understand why, and how the need for fiscal
backing might manifest itself, we first consider a stripped-down model to
illustrate the principles at work.
A representative agent solves
maxC,B,M,F
∫ ∞
0e−βt log(Ct) dt subject to (1)
Ct · (1 + ψ(vt)) +B + M
Pt+ τt + Ft = ρFt +
rtBt
Pt+ Yt , (2)
where C is consumption, B is instantaneous nominal bonds paying interest
at the rate r, M is non-interest-bearing money, ρ is a real rate of return on
a real asset F, Y is endowment income, and τ is the primary surplus (or
simply lump-sum taxes, since we have no explicit government spending in
this model). Velocity vt is given by
vt =PtCt
Mt, vt ≥ 0, (3)
and the function ψ(.), ψ′(.) > 0 captures transaction costs.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 3
The government budget constraint is
B + MPt
+ τt =rtBt
Pt. (4)
Monetary policy is an interest-smoothing Taylor rule:
r = θr ·(
r + θπ
( PP− π
)− r)
. (5)
The “Taylor Principle”, that θπ should exceed one, is the usual prescription
for “active” monetary policy.
First order conditions for the private agent are
∂C :1C
= λ(1 + ψ + ψ′v) (6)
∂F : − ˆλ = λ(ρ− β) (7)
∂B : −ˆλP+ β
λ
P+
λ
P
ˆPP= r
λ
P(8)
∂M : −ˆλP+ β
λ
P+
λ
P
ˆPP= ψ′v2 (9)
The ˆzt notation means the time derivative of the future expected path of z
at t. It exists even at dates when z has taken a jump, so long as its future
path is right-differentiable. Below we also use the d+dt operator for the same
concept.
We are taking the real rate ρ as exogenous, and in this simple version of
the model constant. The economy is therefore being modeled as either hav-
ing a constant-returns-to scale investment technology or as having access to
international borrowing and lending at a fixed rate. Though we could ex-
tend the model to consider stochastically evolving Y, ρ, and other external
disturbances, here we consider only surprise shifts at the t = 0 starting date,
with perfect-foresight deterministic paths thereafter. This makes it easier
to follow the logic, though it makes the time-0 adjustments unrealistically
abrupt.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 4
Besides the exogenous influences that already appear explicitly in the
system above (ρ and Y), we consider an “inflation scare” variable x. This
enters the agents’ first order condition as a perturbation to inflation expec-
tations. It can be reconciled with rational expectations by supposing that
agents think there is a possibility of discontinuous jumps in the price level,
with these jumps arriving as a Poisson process with a fixed rate. This would
happen if at these jump dates monetary policy created discontinuous jumps
in M. Such jumps would create temporary declines in the real value of gov-
ernment debt B/P which might explain why such jumps are perceived as
possible. If the jump process doesn’t change after a jump occurs, there is no
change in velocity, the inflation rate, consumption, or the interest rate at the
jump dates. Rather than solve a model that includes such jumps, we model
one in which the public is wrong about this — there are no jumps, despite
the expectation that there could be jumps. After a long enough period with
no jumps, the public would probably change its expectations, but there is
no logical contradiction in supposing that for a moderate amount of time
the fact that there are no jumps does not change expectations. In fact, if we
consider time-varying paths for x, in which x returns to zero after some pe-
riod, there is no way to distinguish whether the “true” model is one with
the assumed x path (and thus a non-zero probability of jumps in P) or one
with x ≡ 0 if jumps do not actually occur.
The inflation scare variable changes the first order conditions above to
give us
∂B : − ˆλ1P+ β
λ
P+
λ
P
(ˆPP+ x
)= r
λ
P(8’)
∂M : − ˆλ1P+ β
λ
P+
λ
P
(ˆPP+ x
)= ψ′v2 (9’)
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 5
Because the price and money jumps have no effect on interest rates or con-
sumption, no other equations in the model need change. These first or-
der conditions reflect the private agents’ use of a probability model that
includes jumps in evaluating their objective function.
We can solve the model analytically to see the impact of an unantici-
pated, permanent, time-0 shift in ρ (the real rate of return), x (the inflation
scare variable), or ρ (the central bank’s interest-rate target). We also solve
an expanded version below numerically for given exogenous time paths of
those variables.
Solving to eliminate the Lagrange multipliers from the first order condi-
tions we obtain
ρ = r−ˆPP− x (10)
r = ψ′(v)v2 (11)
− d+
dt
(1
C · (1 + ψ + ψ′v)
)=
ρ− β
C(1 + ψ + ψ′v). (12)
Using (10) and the policy rule (5), we obtain that along the path after the
initial date,
r = θr · ((θπ(r− ρ− x)− r + r− θππ)
= θr · (θπ − 1)r− θrθπ(ρ + x) + θr(r− θππ) . (13)
With the usual assumption of active monetary policy, θπ > 1, so this is an
unstable differential equation in the single endogenous variable r. Solutions
are of the form
rt = Et
[∫ ∞
0e−(θπ−1)θrsθrθπ(ρt+s + xt+s) ds
]− r− θππ
θπ − 1+ κe(θπ−1)θrt. (14)
In a steady state with x and ρ constant (and κ = 0), this give us
r =θπ(ρ + x)
θπ − 1− r− θππ
θπ − 1. (15)
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 6
From (11) we can find v as a function of r. Substituting the government bud-
get constraint into the private budget constraint gives us the social resource
constraint
C · (1 + ψ(v)) + F = ρF + Y. (16)
Solving this unstable differential equation forward gives us
Ft = Et
[∫ ∞
0exp
(−∫ s
0ρt+v dv
)(Yt+s − Ct+s(1 + ψ(vt+s))) ds
]. (17)
Here we do not include an exponentially explosive term because that would
be ruled out by transversality in the agent’s problem and by a lower bound
on F. With constant ρ, x and Y, r and therefore v are constant, and (12)
therefore lets us conclude that C grows (or shrinks) steadily at the rate ρ− β.
We can therefore use (17) to conclude that along the solution path, since ρ,
Y and v are constant
Ct =β · (ρ−1Y + Ft)
1 + ψ(v). (18)
This lets us determine initial C0 from the F0 at that date. From then on Ct
grows or shrinks at the rate ρ − β and the resulting saving or dissaving
determines the path of Ft from (18).
III. UNSTABLE PATHS, UNIQUENESS, FISCAL BACKING
To this point we have not introduced a central bank balance sheet or
budget constraint. We can nonetheless distinguish between monetary pol-
icy, controlling the interest rate or the money stock, and fiscal policy, con-
trolling the level of primary surpluses. Passive fiscal policies make the pri-
mary surplus co-move positively with the level of real debt and guarantee
a stable level of real debt, regardless of the time path of prices, under the as-
sumption of stable real interest rates. Passive fiscal policies generally leave
the price level indeterminate, no matter what interest rate or money stock
policy is in place. Guaranteeing uniqueness of the price level requires a
commitment to active fiscal policy. Nonetheless fiscal policy in the presence
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 7
of low inflation may be passive, so long as it is believed that policy would
turn active if necessary to rule out explosive inflation. In the remainder of
this section we make these points analytically in the simple model.
Our solution for r, given by (14), tells us that r could be constant, but
nothing in the model to this point tells us that κ 6= 0 is impossible. To
assess whether these paths are potential equilibria in the model, we need
to specify fiscal policy. The standard sort of fiscal policy to accompany the
type of monetary policy we have postulated (Taylor rule with θπ > 1) is
a “passive” policy that makes primary surpluses plus seigniorage respond
positively to the level of real debt. For example, we can assume
MP
+ τ = −φ0 + φ1BP
. (19)
Substituting this into the government budget constraint (4) and using (10)
gives us
b =
(ρ + x +
ˆP− PP− φ1
)b + φ0 . (20)
On an equilibrium path,ˆPP=
PP
,
that is, actual inflation and model-based expected inflation are equal. Thus
if φ1 > ρ + x, this is a stable differential equation, with b converging to
φ0/(φ1 − ρ − x). In fact, any φ1 > 0 is consistent with equilibrium, even
though for small values b grows exponentially. The transversality condition
with respect to debt for the private agent who holds the debt is
E0
[e−βt λB
Pt
]= 0 . (21)
From (7) λ grows at the rate β− ρ, while from (20) b grows asymptotically
at ρ + x− φ1. However the E0 in the transversality condition is the private
agent’s expectation operator. Since the agent believes in the possibility of
price jumps, the agent thinks that the expected real return on real debt is
just ρ, not ρ + x. Thus the agent believes that b grows asymptotically at the
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 8
rate ρ − φ1. The agent’s transversality condition is therefore satisfied for
any φ1 > 0. The agent in such equilibria has ever-growing wealth, but at
the same time ever-growing taxes that offset that wealth, so that the agent
is content with the consumption path defined by the economy’s real equi-
librium.4
A passive fiscal policy with φ1 > 0, therefore, guarantees that all con-
ditions for a private agent optimum are met on any of the paths for prices
and interest rates we have derived, including those with κ > 0. The infla-
tion rate (not just the price level) diverges to infinity on such a path, along
with the interest rate and velocity. So long as r is an increasing function
of v (ψ′′(v)v2 + 2vψ′(v) > 0), real balances shrink on these paths and, de-
pending on the specification of the ψ(v) function, may go to zero in finite
time.
With κ < 0, the initial interest rate and inflation rate are below the level
consistent with stable inflation and both the price level and the interest rate
decline on an exponential path. Since negative nominal interest rates are
not possible, it is impossible to maintain the Taylor rule when it prescribes,
as it eventually must on such a path, negative interest rates. The simplest
modification of the policy rule that accounts for this zero bound on the in-
terest rate, has r follow the right-hand side of (5) whenever this is positive
or r itself is positive, and otherwise sets r to zero. With this specification and
the passive fiscal rule (19) the economy has a second steady state (assuming
φ1 large enough to stabilize b), at r = 0, b = φ0/(φ1 − ρ− x). In this steady
state inflation is constant at −ρ− x. This steady state is stable.
4Note that, because the realized real rate of return on debt exceeds that on real assets F,
the properly discounted present value of future taxes exceeds the real value of debt on a
path with x > 0, and may even be infinite. “Ricardian” fiscal policy does not guarantee a
match between the present value of future taxes and the current real value of debt on this
non-rational-expectations path for the economy.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 9
At this point we have approximately matched the model and conclu-
sions of Benhabib, Schmitt-Grohé, and Uribe (2001): This policy configura-
tion produces a pair of equilibria, with only one globally stable. Because the
equilibria with κ > 0 cannot be ruled out, and because there are many paths
for the economy that converge to the stable r = 0 point, the price level is
indeterminate.
However, the indeterminacy can be eliminated if we specify plausible
modifications of the policy rules for very high and low inflation rates. The
passive fiscal rule, when r and inflation are on an upward-explosive path,
requires that the conventional deficit, which includes interest expense, ex-
plode upward asymptotically at the same rate as r. This is required because
inflation is tending to reduce the real value of the debt, so large amounts of
additional nominal debt must be sold to the public to keep the real value of
the debt on its path converging to φ0/(φ1− ρ− x). This behavior of the pol-
icy authorities is implausible. It is natural to suppose that at very high infla-
tion rates the fiscal authorities would try to restrain the conventional deficit
by increasing τ, and that the monetary authorities would try to refrain from
exacerbating the conventional deficit by continuing to push r upward. In
Sims (forthcoming) it is shown in a model with only interest-bearing debt,
no currency or transactions costs, that even a tiny positive coefficient on in-
flation in the fiscal policy rule would make the explosive paths for inflation
unsustainable. In such a model it is also straightforward to describe policies
that differ from the standard passive active money, passive fiscal rule only
far from steady state and that also deliver a unique equilibrium price level.
In a model like that we use here, with transactions costs and non-interest-
bearing currency, the details of a modest policy shift at high or low inflation
rates that would guarantee a unique equilibrium price level depend on the
way transactions costs behave at very high and very low levels of veloc-
ity. Though the details might be complex, the essence of such a backup
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 10
policy is simple. If inflation gets too high, a modest fiscal and monetary
reform is undertaken that “punishes” market participants who have been
expecting ever-accelerating inflation by suddenly, but moderately, increas-
ing the value of the currency. If market participants see from the start that
this will happen, the explosive equilibrium path can never get started. If
market participants are not so far-sighted, the economy might start down
such a path, but as soon as people realized where the economy was headed
market forces would restore the stable-price equilibrium.
To eliminate the solution paths that converge to the low-inflation steady
state, we can invoke a different sort of realistic modification of the simple
active-money, passive-fiscal policy rules. The version of passive fiscal pol-
icy in Benhabib, Schmitt-Grohé, and Uribe (2001) makes the primary sur-
plus respond positively to the real value of all government liabilities, both
interest-bearing and zero-interest. But this makes little sense. There is no
need for taxes to increase with the real value of currency. In our model,
as the economy approaches the zero lower bound on nominal interest rates
(which it does in finite time on these deflationary paths if the standard Tay-
lor rule remains in place), real balances increase without bound. (v → 0 as
r → 0 and C does not decrease.) So long as these increased real balances
are not offset by correspondingly increased taxation and correspondingly
increased net lending by the government, they make the market value of
private wealth increase without bound in finite time. This violates the pri-
vate sector transversality condition. Less technically, individuals will not
be content with consumption satisfying the economy’s resource constraint
if the market value of their wealth grows arbitrarily large.
We can conclude that there is no internal contradiction in the conven-
tional practice of treating the price level as uniquely determined in models
with a Taylor rule. The justification for doing so, though, must appeal to a
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 11
backstop fiscal policy commitment — to tighten fiscal policy if inflation be-
comes too high, and to allow the real value of currency to increase without
bound, without raising taxes, if deflation takes hold. Central banks should
therefore not be structured to have no institutional link to the treasury, and
central bankers should not suggest in their public statements that they can
control inflation regardless of fiscal policy.
IV. FOUR LEVELS OF CENTRAL BANK BALANCE SHEET PROBLEMS
So far, we have said nothing about the central bank balance sheet, but
with the solution path for the economy in hand, assessing the time path of
the balance sheet is straightforward. The most severe problem, which we
can call level 4, is simply the possible indeterminacy of the price level. To
put this in the language of the central bank balance sheet, this is the point
that the central bank’s assets consist of the market value of its assets and its
potential seigniorage, both of which are valueless if currency is valueless.
But if it holds nominal debt as assets and issues reserves and currency as
liabilities, the central bank has no lever to guarantee the real value of either
side of its balance sheet. If the public were to cease to accept currency in
payment, it would become valueless, as would both sides of the central
bank balance sheet. That this cannot happen, either suddenly or as the end
point of a dynamic process, depends on fiscal commitments beyond the
central bank’s control.
The fiscal backing required for price level determinacy seems quite plau-
sible in the US. In Europe, because fiscal responsibility for the Euro is di-
vided among many countries that seem bent on frequently increasing doubts
about their ability to cooperate on fiscal matters, the possibility of a break-
down of the value of the Euro from this source cannot be entirely ruled out.
The next level of possible problem, level 3, arises because the notion
of determinacy via a backstop fiscal commitment assumes that the central
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 12
bank could maintain its commitment to an active policy rule during an in-
flationary excursion from the unique stable price path, up to the point that
fiscal backing is triggered. If we think of a unified government budget con-
straint and jointly determined monetary and fiscal policy, this is not an is-
sue. But if the central bank is concerned to maintain its policies without
requiring a direct capital injection from the treasury, or possibly even with-
out ever having to set its seigniorage payments to the treasury to zero, then
this could be a problem. And of course if markets perceive that the central
bank will abandon its policy rule to avoid having to seek treasury support,
this undercuts the argument for price determinacy. Showing formally how
these issues arise requires solving the model for time varying paths of inter-
est rates and velocity, so it is postponed to later sections of the paper.
If the market value of the assets of the central bank fall to a value be-
low that of their interest-bearing liabilities, it is possible that adherence to
the bank’s policy rule is impossible without a direct injection of capital.
This is only a possibility, however, because the bank has an implicit asset
in its future seigniorage. Even with assets below interest-bearing liabili-
ties at market value, the bank may be able to meet all its interest-paying
obligations and to restore the asset side of its balance sheet through accu-
mulation of seigniorage. Whether it can do so depends on its policy rule
and on the interest-elasticity of demand for currency (or more generally, for
its non-interest-bearing liabilities). This issue, of whether the central bank
might require a capital injection to maintain adherence to its policy rule in
a determinate-price-level equilibrium, is a level 2 balance sheet problem.
Finally, the central bank may be solvent in the sense that with the ex-
isting policy rule its assets at market value plus future seigniorage exceed
its total liabilities, yet following standard accounting rules and rules for de-
termining how much seigniorage revenue is sent to the treasury each pe-
riod may lead to episodes of zero seigniorage payments to the treasury.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 13
Extended episodes of this type might be thought to raise issues of politi-
cal economy, if they led to public criticism of the central bank or to calls for
revising its governance.
V. INFLATION SCARE IN THE SIMPLE MODEL
Our first numerical example uses this simple model to compare a steady
state with ρ = β = ρ = .01 and x = 0 to one in which x jumps up to .02 at
time 0. This is an “inflation scare” scenario. The 2% per year inflation scare
shock produces a much larger increase in the nominal interest rate, because
the increased inflation expectations shrink demand for money and thereby
produce inflation, which prompts the central bank to raise rates further. If
the duration of the nominal assets on the central bank’s balance sheet is
positive, the permanent rise in rates reduces the time 0 market value of the
central bank’s assets. The simple model treats the debt as of maturity 0, but
this has no consequence except for the initial date capital losses, because for
t > 0 the perfect-foresight path requires that long and short debt has the
same time path of returns. We show two cases: initial assets of the central
bank A0 are three times the amount of currency outstanding or six times
the amount of currency outstanding with the initial deposit liabilities V0
plus currency matching A0 in each case.
The nominal capital losses, as a proportion of the new value of the assets,
are shown in Table 1. There cannot be any “level 2” problem for the central
bank unless the interest increase pushes its initial assets A below V. That
is, it not only has to have assets less than liabilities V + M, where M is
currency, it has to have A < V in order for a level 2 problem to arise. The
rise in interest rate reduces the demand for M, which has to be met either
by an decrease in A through open market sales or an increase in V. This
will dampen the effect on V − A of the rate rise. The value of V − B is
shown as the “gap” line in Table 1. Whether a level 2 problem actually arises
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 14
then depends on the discounted present value of the seigniorage after the
initial date, shown as “dpvs” in the table. For this example, even though
the gap between V and B gets quite large if we assume long durations for
the assets, the gap exceeds the discounted present value of seigniorage only
for durations of 10 years or more and for the (unrealistically large) balance
sheet with A0 six times outstanding currency.
This example should make it clear that the central bank can suffer very
substantial capital losses without needing direct recapitalization. On the
other hand, it shows that there are drawbacks to extreme expansion of cen-
tral bank holdings of long-maturity debt — an expanded balance sheet in-
creases the probability that interest rate changes could require a direct cap-
ital injection.
This simple model has omitted two sources of seigniorage, population
growth and technical progress, and has considered only a single, stylized,
shock to the balance sheet. We now expand the model to include these ex-
tra elements and calibrate the parameters and the nature of the shocks more
carefully to the situation of the US Federal Reserve. Of course our ability to
calibrate is limited by the sensitivity of results to the transactions cost func-
tion. We have little relevant historical experience with currency demand
at low or very high interest rates. Rates were very low in the 1930’s and
the early 1950’s, but the technology for making non-currency transactions
is very different now. It is difficult to predict how much and how fast people
would shift toward, say, interest-bearing pre-loaded cash cards as currency
replacements if interest rates increased to historically normal levels. We can
at best show ranges of plausible results.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 15
VI. THE MODEL
The model borrows from Sims (2005). The household planner (whose
utility includes that of offspring, see Barro and Sala-i Martin (2004)) maxi-
mizes: ∫ ∞
0e−(β−n)t log(Ct)dt (22)
where Ct is per capita consumption, β is the discount rate, and n is popula-
tion growth, subject to the budget constraint:
Ct(1 + ψ(vt)) + Ft +Vt + Mt + qtBP
Pt=
Yeγt + (ρt − n)Ft + (rt − n)Vt
Pt+ (χ + δ− qtδ− n)
BP
P− n
Mt
Pt− τt. (23)
We express all variables in per-capita terms and initial population is normal-
ized to one. Ft and BPt are foreign assets and long-term government bonds
in the hand of the public, respectively, Vt denotes central bank reserves, Mt
is currency, τt is lump-sum taxes, Y is an exogenous income stream growing
at rate γ. Foreign assets and central bank reserves pay an exogenous real
return ρ and a nominal return rt, respectively. Long term bonds are mod-
eled as in Woodford (2001). They are assumed to depreciate at rate δ (δ−1
captures the bonds average maturity) and pay a nominal coupon χ + δ.5
The government is divided into two distinct agencies called “central
bank” and “fiscal authority”. The central bank’s budget constraint is(qt
BC
Pt− Vt + Mt
Pt
)ent
=
((χ + δ− δqt − nqt)
BC
P− (rt − n)
Vt
Pt+ n
Mt
Pt− τC
t
)ent. (24)
where BCt are long-term government bonds owned by the central bank, and
τCt are remittances from the central bank to the fiscal authority. The central
5We write the coupon as χ + δ so that at steady state if χ equals the short term rate the
bonds sell at par.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 16
bank is assumed to follow the rule (5) for setting rt, the interest on reserves.
The central bank is also assumed to follow a rule for remittances, which
embodies two principles: i) remittances cannot be negative, ii) whenever
positive, remittances are such that the central bank capital (assets minus li-
abilities) remains constant in nominal terms over time (Hall and Reis (2013)
use a similar rule), that is:(qtBC
t −Vt −Mt
)ent = constant. (25)
Differentiating the condition above and plugging the resulting expression
into the central bank’s budget constraint, one can see that the two principles
result in the following rule for remittances:
τCt = max
{0, (χ + (1− qt)δ + qt)
BCt
Pt− rt
Vt
Pt
}. (26)
Solving the central bank’s budget constraint forward we can obtain its in-
tertemporal budget constraint:
qBC
0P0− V0
P0+∫ ∞
0(
Mt
Mt+ n)
Mt
Pte−∫ t
0 (ρs−n)dsdt =∫ ∞
0τC
t e−∫ t
0 (ρs−n)dsdt. (27)
Equation (27) shows that, regardless of the rule for remittances, their dis-
counted present value∫ ∞
0τC
t e−∫ t
0 (ρs−n)dsdt has to equal its left hand side,
namely the market value of assets minus reserves plus the discounted present
value of seigniorage∫ ∞
0(
Mt
Mt+ n)
Mt
Pte−∫ t
0 (ρs−n)dsdt. We can also compute
the constant level of remittances τCeγt (taking productivity growth into ac-
count) that satisfies expression (27).
τC =
(∫ ∞
0e(γ+n)t−
∫ t0 ρsdsdt
)−1
(q
Bc
P− V
P+∫ ∞
0(
Mt
Mt+ n)
Mt
Pte−∫ t
0 (ρs−n)dsdt)
. (28)
We also need to specify the central bank’s policy in terms of the asset
side of its balance sheet BCt . We assume that these follow some exogenous
process BCt = BC
t . Government debt is assumed to be held either by the
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 17
central bank or the public: Bt = BCt + BP
t . The budget constraint of the fiscal
authority is(Gt − τt + (χ + δ− δqt − nqt)
BP
)ent =
(τC
t + qtBPt
)ent, (29)
where Gt is government spending. The rule for τt is given by:
τt = ξ0eγt + (ξ1 + n + γ)
(q
BP
P+
VP
). (30)
This rule makes the debt to GDP ratio bt =
(q
BP
P+
VP
)e−γt converge as
long as ξ1 > β− n. The initial level of foreign assets in the hand of the pub-
lic, central bank reserves, and currency are FP0 , V0, and M0, respectively.
As in the simple model the first order condition for the household’s
problem with respect to C, FP, B, V, and M yield the Euler equation (12), the
Fisher equation (10),6 the money demand equation (11), and the arbitrage
condition between reserves and long-term bonds:
χ + δ
q− δ +
qq= r. (31)
The solutions for r is given by equation (14), and those for inflationPP
and
velocity v follow from equations (10) and (11), respectively. The growth rate
of consumptionCC
, is given by
CC
= (ρ− β)− 2ψ′(v) + vψ′′(v)1 + ψ(v) + vψ′(v)
v, (32)
which obtains from differentiating expression (12). Differentiating the def-
inition of velocity (3) we obtain an expression for the growth rate of cur-
rency:MM
=PP+
CC− v
v. (33)
6Note that short term debt was called B in the simple model, and was issued by the
fiscal authority. Here it is called V, and is issued by the central bank.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 18
The economy’s resource constraint is given by
C(1 + ψ(v)) + F = (Y− G)eγt + (ρ− n)F, (34)
where F = FP + FC is the aggregate amount of foreign assets held in the
economy (we assume that the central banks foreign reserves FC are zero),
and where we assumed Gt = Geγt. Solving this equation forward we obtain
a solution for consumption in the initial period:
C0
(∫ ∞
0(1 + ψ(v))e−
∫ t0 (ρs− C
C−n)dsdt)= F0 + (Y− g)
∫ ∞
0e(γ+n)t−
∫ t0 ρsdsdt,
(35)
Given velocity v and the level of consumption, we can compute real money
balancesMP
, the initial price level P0, and seigniorageMP
+nMP
=
(MM
+ n)
MP
(using (33)), and the present discounted value of seigniorage
∫ ∞
0
(MM
+ n)
MP
e−∫ t
0 (ρs−n)dsdt = c0
∫ ∞
0
(MM
+ n)
v−1e−∫ t
0 (ρs− CC−n)dsdt.
Finally, solving (31) forward we find the current nominal value of long-term
bonds
q0 = (χ + δ)∫ ∞
0e−(∫ t
0 rsds+δt)
dt. (36)
VI.1. Steady state. At a steady state where ρ = β + γ, r = ρ + π, v satis-
fies v2ψ′(v) = rss. Steady state consumption is given by Ct = C0eγt where
C0 =(β− n)F0 + Y− G
1 + ψ(v), and real money balances are given by
MP ss =
C0
veγt. seigniorage is given by (π + γ + n)
C0
ve(γ+n)t and its present dis-
counted value is given by (π + γ + n)C0
v(β− n).
VI.2. Central bank’s solvency, accounting, and the rule for remittances.
For some of the papers discussed in the introduction the issue of central
bank’s solvency is simply not taken into consideration: the worst that can
happen is that the fiscal authority may face an uneven path of remittances,
with possibly no remittances at all for an extended period. We acknowledge
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 19
the possibility that remittances may have to be negative, at least at some
point. This is what we mean by solvency.
Like Bassetto and Messer (2013), we approach the issue of central bank’s
solvency from a present discounted value perspective. If the left hand side
of equation (27) is negative, the central bank cannot face its obligations, i.e.,
pay back reserves, without the support of the fiscal authority. An inter-
esting aspect of equation (27) is that its left hand side does not depend on
many of aspects of central bank policy that are recurrent in debates about
the fiscal consequences of central bank’s balance sheet policy. For instance,
the future path of BCt does not enter this equation: whether the central bank
holds its assets to maturity or not, for instance, is irrelevant from an ex-
pected present value perspective. Intuitively, the current price qt contains
all relevant information about the future income from the asset relative to
the opportunity cost rt. Whether the central bank decides to sell the assets
and realize gains or losses, or keep the assets in its portfolio and finance
it via reserves, does not matter. Similarly, whether the central bank incurs
negative income in any given period, and accumulates a “deferred asset”,
is irrelevant from the perspective of the overall present discounted value of
resources transferred to the fiscal authority.7 In fact, we will see later that in
some cases scenarios associated with higher remittances in terms of present
value are also associated with a deferred asset.
Finally, the issue of “remittances smoothing” is also, from a purely eco-
nomic point of view, a non issue. In perfect foresight the central bank
can always choose a perfectly smooth path of remittances (in fact, this is
τCt = τCeγt). In a stochastic environment Barro’s results on tax smooth-
ing would apply: remittances would move with innovations to the the left7As we will see later central bank accounting does not let negative income affect capital.
The budget constraint (24) implies however that negative income results in either more
liabilities or less assets. In order to maintain capital intact, a deferred asset is therefore
created on the asset side of the balance sheet.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 20
hand side of equation (27). But there are accounting rules governing central
banks’ remittances. Hence these may not be smooth and may depend on
the central bank’s actions, such as holding the assets to maturity or not. We
recognize that the timing of remittances can matter for a variety of reasons:
tax smoothing, political pressures on the central bank, et cetera. For this
reason we assume a specific rule for remittances that very loosely matches
those adopted by actual central banks and compute simulated paths of re-
mittances under different assumptions. Appendix A discusses this rule.
VI.3. Functional forms and parameters. Table 2 shows the model parame-
ters. We normalize Y − g to be equal to 1, and set F0 to 0.8 Since we do not
have investment in our model, and F0 = 0, Y− G in the model corresponds
to national income Y minus government spending G in the data (data are
from Haver analytics, mnemonics are Y@USNA and G@USNA, respectively).
All real quantities discussed in the remainder of the paper should therefore
be understood as multiples of Y− G, and their data counterparts are going
to be expressed as a fraction of national income minus government spend-
ing ($ 11492 bn in 2013Q3). Our t = 0 corresponds to the beginning of 2014.
We therefore measure our starting values for the face value of central bank
assetsBC
P, reserves
VP
, and currencyMP
using the January 3, 2014 H.4.1 re-
port (http://www.federalreserve.gov/releases/h41/), which mea-
sures the Security Open Market Account (SOMA) assets.9 The model pa-
rameters are chosen as follows. The discount rate β, productivity growth γ,
8Note from the steady state calculations that we could choose F0 6= 0 and use instead
the normalization (β − n)F0 + y − g=1, hence setting F0 6= 0 simply implies a different
normalization.9The January 3, 2014 H.4.1 reports the face value of Treasury ($ 2208.791 bn ), GSE debt
securities ($ 57.221 bn), and Federal Agency and GSE MBS ($ 1490.160 bn), implying that
BC0 is $ 3756.172 bn, the value of reserves V (deposits of depository institutions, $ 2374.633
bn) and currency M (Federal Reserve notes outstanding, net, $ 1194.969 bn).
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 21
and population growth n are 1 percent, 1 percent, and .75 percent, respec-
tively. These values are consistent with Carpenter et al.’s assumptions of a
2% steady state real rate.
The policy rule has inflation and interest rate smoothing coefficients θπ
and θr of 2 and 1, respectively, which are roughly consistent with those of in-
terest feedback rules in estimated DSGE models (e.g., Del Negro, Schorfheide,
Smets, and Wouters (2007); note that θr = 1 corresponds to an interest rate
smoothing coefficient of .78 for a policy rule estimated with quarterly data).
The inflation target θπ is 2 percent (hence θ0 = β + γ− (θπ − 1)πss=-.0025).
We use the functional form
ψ(v) = ψ0v
1 + ψ1v(37)
for the transaction costs, with ψ0 = 2× 10−6 and ψ1 = −0.055. Figure 1
shows the scatter plot of quarterlyMPC
= v−1 and the annualized 3-month
TBill rate in the data (where M is currency and PC is measured by nom-
inal PCE)10, where blue crosses are post-1980 data, and crosses are 1947-
1980 data (arguably less relevant). The black curve in figure 1 shows the
relationship between velocity and interest rates as implied by the model
(equation (11)). The parameters ψ0 and ψ1 are chosen to i) match currency
demand in real termsMP
in 2013Q4 at current rates (r0=.0025),11 ii) so thatMPC
= v−1 asymptotes at ψ1 = .055 when rates go to infinity, which implies
that model-implied velocity is roughly in line with the experience of the
early 1980s, as shown in figure 1. The implied transaction costs at steady
state are negligible - about .03 percent of Y-G. We will later also consider
10Data are from Haver, with mnemonics C@USNA, FMCN@USECON, and FTBS3@USECON
for PCE, currency, and the Tbill rate, respectively.11Matching inverse velocity
MPC
in 2013Q4 as opposed to real money demandMP
yields
very similar results.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 22
alternative parameterizations of currency demand.12 Finally, we choose
χ – the average coupon on the central bank’s assets – to be 3.5 percent,
roughly in line with the numbers reported in figure 6 of Carpenter, Ihrig,
Klee, Quinn, and Boote (2013). Chart 17 of the April 2013 FRBNY report
on “Domestic Open Market Operations during 2012”13 shows an average
duration of 6 years for SOMA assets (SOMA is the System Open Market
Account, which represents the vast majority of the Federal Reserve balance
sheet). We therefore set 1/δ =6.
VII. SIMULATIONS
As a baseline simulation we choose a time-varying path of short term
nominal interest rates that roughly corresponds to the baseline interest path
in Carpenter, Ihrig, Klee, Quinn, and Boote (2013). We generate this path by
assuming that the real rate ρt remains at a low level ρ0 for a period of time
T0 equal to five years, and then reverts to the steady state ρ at the rate ϕ1:
ρt =
ρ0, for t ∈ [0, T0]
ρ + (ρ0 − ρ)e−ϕ1(t−T0), for t > T0.(38)
Given the path for ρt, equation (14) generates the path for the nominal short
term rate (we set κ = 0 for the baseline simulation). The baseline paths of
ρt, rt and inflation πt are shown as the solid black lines in the three panels
of Figure 2.
12We have performed non-linear least squares regression of equation (11) using the
interest rate and velocity data shown in figure 1. Some of the estimates of ψ0 and ψ1 –
particularly those using post-1980 data – are quite close to those reported in table 2. These
estimates generally yield a value of ψ1 close to −.06 in order to fit the high inflation data of
the early 1980s, which we as implying too large an asymptote forMPC
= v−1 if rates were
to become very large, in light of current transaction technology.13http://www.newyorkfed.org/markets/omo/omo2012.pdf.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 23
Given the path for ρt and rt we can compute q and the amount of re-
sources, both in terms of marketable assets and present value of future
seigniorage, in the hands of the central bank. The first row of table 3 shows
the two components of the left hand side of equation (27), namely the mar-
ket value of assets minus reserves (column 1) and the discounted present
value of seigniorage∫ ∞
0(
MM
+ n)MP
e∫ t
0 (ρs−n)dsdt (column 2). The third col-
umn shows the sum of the two, which has to equal the discounted present
value of remittances∫ ∞
0τCe
∫ t0 (ρs−n)dsdt. Column 4 shows τC as defined in
equation (28): the constant level of remittances (accounting for the trend in
productivity) that would satisfy equation (27), expressed as a fraction of Y-
G like all other real variables.14 Last, in order to provide information about
how the numbers in column 1 are constructed, column 5 shows the nominal
price of long term bonds q at time 0.
Under the baseline the real value of the central bank’s assets minus lia-
bilities is 14.6 percent of Y-G – which is larger than the difference between
the par value of assets minus reserves reported in table 2 given that q is
above one under the baseline. Its value is 1.08, which is above the 1.04 ra-
tio of market over par value of assets reported in Federal Reserve System
(2014)15 The discounted present value of seigniorage is almost an order of
magnitude larger, however, at about 99 percent of Y-G, and represents the
bulk of the central bank resources (and therefore the present discounted
value of remittances), which are 114 percent of Y-G. The constant (in pro-
ductivity units) level of remittances τC that satisfies the present value rela-
tionship is .26 percent of Y-G, about $ 29 bn per year, quite lower than the
amount remitted for 2013 and 2012 according to Federal Reserve System
(2014) ($ 79.6 and $ 88.4 bn, respectively).
14That is, the amount τC such that τCt = τCeγt satisfies the present value relationship.
15Page 23 and 29 shows the par and market (fair) value of Treasury and GSE debt secu-
rities, and Federal Agency and GSE MBS, respectively.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 24
The left and right panels of Figures 3 show inverse velocity M/PC and
seigniorage, expressed as a fraction of Y-G, in the data (1980-2013) and in
the model (under the baseline simulation), respectively. A comparison of
the two figures shows that the drop in M/PC as interest rates renormalize
under the baseline simulation (from about .085 to .065, left axis) is roughly
as large as the rise in M/PC as interest rates fell from 2008 to 2013. Partly
because the model may likely over-predict the fall in currency demand, and
more importantly because consumption declines (real interest rates are very
low at time 0, inducing unrealistic above trend consumption), seigniorage
falls to negative territory for roughly six years. After that, it converges to
almost .3 percent of Y-G, a level that is in the low range of the post-1980
observations. For both reasons the present discounted value of seigniorage
reported in table 2 for the baseline simulation is likely to be a fairly conser-
vative estimate.
Finally, the left panel of Figure 4 shows remittances (computed as de-
scribed in section A) under two scenarios for the path of assets BC: in the
first scenario (solid line) the central bank lets its assets depreciate, while in
the second one it actively sells assets at a rate of 20 percent per year. These
scenarios highlight the fact that different paths for the balance sheet can
imply different paths for remittances, even though their expected present
value remains the same (this is the dotted line in figure 4, which shows
τCeγt).
Next, we consider alternative simulations where the economy is subject
to different “shocks.” In each of these simulations all uncertainty is revealed
at time 0, at which point the private sector will change its consumption
and portfolio decisions and prices will adjust. We will use the subscript
0− to refer to the pre-shocks quantities and prices (that is, the time 0 quan-
tities and prices under the baseline simulation). For each simulation, Ta-
ble 3 will report the new market value of assets minus reserves in real term
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 25
(q0BC
0−
P0− V0
P0). By assumption the central bank will not change its assets BC
0−
after the new information is revealed, but the private sector will change its
time 0 currency holdings given that interest rates may have changed. This
necessarily leads to a change in reserves (given that central bank’s assets are
unchanged) equal toV0 −V0−
P0= −M0 −M0−
P0in real terms. We report this
quantity separately in column 5.
Last, for each scenario we also report the level of the balance sheet BC
such that, for any balance sheet size larger than BC, the present discounted
value of remittances (see equation (27)) becomes negative after the shock.
We refer to this situation as the central bank becoming “insolvent”, in the
sense that at some point it will need resources from the fiscal authority.
Specifically, assume the central bank expands its balance sheet by ∆BC at
time 0− (right before the shock takes place) by buying assets at price q0−
and pays with it by expanding reserves by an amount ∆V = q0−∆BC. How
large can ∆BC be to still satisfy
q0(
BC + ∆BC)−V − ∆VP0
+M0 −M0−
P0+∫ ∞
0(
Mt
Mt+ n)
Mt
Pte−∫ t
0 (ρs−n)dsdt ≥ 0 (39)
after the “shock”? We report B/BC = 1 +∆BC
BC , where BC is the 2013Q4
level of the balance sheet reported in table 2. Of course, the reason why
with a larger balance sheet the central bank may become insolvent is that q
is lower under the alternative simulations, and hence the central bank may
experience losses, in addition to possibly having less seigniorage in present
value.
The first alternative scenario we consider is a “higher rates” path similar
to one considered by Carpenter, Ihrig, Klee, Quinn, and Boote (2013). Under
this new path real rates converge to a 1 percent higher steady state, and so
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 26
will short term nominal rates given that the central bank inflation target has
not changed. We choose the new starting value for ρ, ρ0, so that the initial
rate remains at 25 basis points. The red solid lines in the three panels of
Figure 2 show the “Higher Rates” paths for the nominal and the real short
term rates and inflation, respectively. In these simulations we assume that
the central bank recognizes the change in the steady state ρ = β + γ, and
adjusts its Taylor rule coefficient r = ρ + π accordingly.
We consider two different reasons why the new steady state ρ is higher:
a higher discount rate β and a higher growth rate of technology γ. While
the new value of q is the same in both cases (the interest rate path is the
same), the present value of seigniorage shown in column 2, and therefore
the present value of remittances shown in column 3, is quite different. In
the high β case the current value of the future income from seigniorage falls
by almost one order of magnitude, as future seigniorage is discounted at
a higher real rate. In the high γ case the economy is growing faster, and
so does money demand and future seigniorage. As it turns out, in both
cases (higher β and higher γ) the level of τC is higher than in the baseline
case. This may seem surprising in the higher β case since the present value
of seigniorage is lower than under the baseline simulation. However, the
central bank is now earning a higher return on its assets, and can therefore
afford a higher level of remittances.16
We also consider the case where the central bank does not recognize that
the steady state real rate has increased, and therefore leaves r unchanged in
its reaction function. This is an extreme version of the more realistic case16Note that our conclusion is different from that reached in the analysis of Carpenter,
Ihrig, Klee, Quinn, and Boote (2013), which take seigniorage as given and focus on the
effect of the higher nominal interest rates on the value of the central bank’s assets qBC,
which falls following the drop in q. The effect of the higher real rate of return on future
central bank’s revenues and, especially in the high γ case, on future seigniorage, trumps in
our simulation the negative effect on q.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 27
where the central bank adjusts its reaction function slowly to changes in
the real economy (see Orphanides (2002)). The red dash-and-dotted lines in
Figure 2 show the paths of nominal interest rates and inflation, respectively,
in this scenario. Inflation is higher relative to the case where the central
bank changes r (solid red lines) because the central bank’s reaction function
calls for rates that are too low. In equilibrium, since inflation is higher, rates
end up being also higher. As a consequence, q falls by more than in the case
where the central bank adjusts r (see the rows labeled “Higher rates/same
intercept” in Table 3). In addition, because of the higher interest rates the
private sector is no longer willing to hold as much currency and turns it into
reserves, and hence a higher fraction of the central bank’s liabilities becomes
interest bearing relative to the baseline scenario. Because of the higher in-
flation, however, given our assumptions about money demand seigniorage
is higher (compare rows 2 and 4 for the higher β case, and 3 and 5 for the
higher γ case, of Table 3). As a consequence, the fall in q does not raise any
solvency issues under these scenarios, which the central bank could have
withstood even with a balance sheet more than five times as large as the
actual one (see column 7). In fact, average yearly remittances τC increase by
about .10 percent of Y-G, roughly $ 10 bn.
The results are very sensitive to the inflation response in the policy re-
action function. The blue dash-and-dotted and dotted blue lines in the top
left panel of Figure 5 show the interest rate path corresponding to an infla-
tion coefficient θπ of 3 and 1.05.17 As is usually the case in stable rational
expectations equilibria, a higher inflation coefficient in the interest rate rule
induces a lower equilibrium response of inflation, and therefore a lower
equilibrium response of interest rates – and vice versa when the inflation
response is lower. In fact, we see that when θπ is 1.05, interest rates reach 25
17In these simulations we change the time 0 real rate so that under the baseline scenario
the nominal rate is still 25 basis points.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 28
percent. Consequently, q falls to less than half its value under the baseline
scenario, and the market value of assets minus reserves q0BC
0−
P0− V0
P0falls to
negative levels (see row 11 of Table 3). The implication of this finding is that
under a large balance sheet the central bank may want to respond more ag-
gressively to inflation, if it is concerned about fluctuations in the values of
its assets.
Even in the θπ = 1.05 case central bank’s solvency is not an issue, how-
ever. The present value of seigniorage has a roughly fivefold increase rela-
tive to the θπ = 2 case, so in spite of the large decline in the value of assets,
the present value of remittances remains positive. In fact, the present value
of remittances would remain positive even if we assumed the central bank
balance sheet to be more than three times as large as the current one (col-
umn 7). We discuss later what would happen in this scenario under a less
favorable outlook for seigniorage (induced by a different parameterization
of the money demand function).
Next, we consider simulations where for a given period of time (10
years) the private sector is concerned about a sudden jump in the price level,
and therefore demands a premium x for holding nominal bonds. We call
these scenarios “inflation scares” and set x to 2 percent. The red solid lines
in the top right and bottom left panels of Figure 5 show what happens to
short term nominal interest rates and inflation, respectively, under this sce-
nario. Because of the higher inflation expectations the central bank, which
follows the interest rule, is forced to raise nominal rates over the baseline
path. This scenario has a number of effects on the central bank’s balance
sheet, as shown in row 6 of Table 3. The market value of assets minus re-
serves q0BC
0−
P0− V0
P0drops to less than half its baseline value, both because q
falls and because the private sector turns currency into reserves. However,
the present discounted value of seigniorage computed under our assump-
tions on money demand is slightly higher than under the baseline case. The
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 29
fact that seigniorage holds up, and that its present discounted value is large,
implies that the present value of resources in the hands of the central bank is
not much affected under the inflation scare simulations. As a consequence,
even with a much larger balances sheet the central bank could have with-
stood the fall in the value of its assets without ever needing any resources
from Fiscal Authority, at least in expectations. Also, because seigniorage
is still large the decrease in average remittances τC is small relative to the
baseline case, from .26 to .24 percent of Y-G.
The blue lines in the top right and bottom left panels of Figures 5 show
the responses of interest rates and inflation, respectively, under policy rules
different from the baseline for this scenario. As in the “Higher rates/same
intercept” case, if policy reacts more aggressively to inflation than under the
baseline in equilibrium inflation and short term interest rise less, and vice
versa if policy reacts less aggressively to inflation. Rows 9 and 12 of Table 3
document the effect of these alternative policies on the central bank’s bal-
ance sheet and show that, again not surprisingly, the effect of the “inflation
scare” simulation on q is stronger the lower the response to inflation θπ.
In fact, under the lower θπ policy, the market value of assets and reserves
becomes negative. The central bank’s overall resources (column 3) are still
sizable, and average remittances are the same as under the θπ = 2 policy.
Again, this is because the higher inflation experienced under the lower θπ
policy yields greater seigniorage (column 2).
Finally, we consider explosive paths where κ in equation (14) is different
from zero. The solid red line in the bottom right panel of Figure 5 shows one
of these paths (with κ = 10−4) under the baseline policy response. Given
the rise in rt under this scenario, q drops by .2 relative to the baseline (row
7 of Table 3). However, q0BC
0−
P0− V0
P0increases to almost four times Y-G.
The reason for this counterintuitive result is that the under our assumption
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 30
for money demand transaction costs explode as rates reach infinity. There-
fore agents will front-load consumption as much as possible while interest
rates and transaction costs are still low, and this will drive up money de-
mand. Moreover, under these explosive paths seigniorage will reach very
high levels because under our parametrization M/Pc (inverse velocity) has
a lower bound, and the public continues to be willing to hold currency no
matter how high rates are. The present discounted value of seigniorage will
therefore be very high (we can show that since seigniorage asymptotes to
a finite level, the present discounted value is actually a finite number, but
is so large that our integration routine does not converge, hence we report
“Inf” in Table 3).
The dash-and-dotted and dotted blue lines in the bottom right panel
of Figure 5 show the responses under different θπ coefficients. In the case
of unstable solutions, the inflation response coefficient in the interest rule
plays the opposite role relative to the stable solution case (see Cochrane
(2011)): the stronger the response, the faster inflation and interest rates ex-
plode. The market value of central bank’s assets q surely falls more with a
higher θπ, but as in the θπ = 2 case the increase in the demand for money
in the initial period and the subsequent large seigniorage imply that central
bank’s solvency is never in question under these paths – quite the contrary,
the central bank transfers very large resources to the fiscal authority.
VII.1. Calibration with money demand equal to zero for r > 100%. We
have seen that under many scenarios where the value of the central bank’s
assets drops substantially, seigniorage can save the day in terms of central
bank’s solvency. This conclusion is not robust to the assumptions about
money demand, however. Table 4 show the central bank’s resources un-
der the baseline, higher rates/same intercept, inflation scare, and explosive
paths simulations for a money demand function that drops to zero when
interest rates are above 100 percent (the other parameter of the transaction
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 31
cost function (37) is still chosen so that currency holdings match 2013Q4
holdings for r = .0025). As shown in the left panel Figure 6, this money de-
mand function predicts an unrealistically large drop in money demand as
interest rates return to steady state, as well as a too low steady state level of
seigniorage. Under the baseline simulation, the present value of seigniorage
is therefore below a third of its value under the benchmark money demand
function. But it does have the arguably realistic feature that for very high
rates of inflation the private sector will use means of transactions other than
cash. As a consequence under the explosive path there is no time 0 jump
in money demand (people are no longer twisting their consumption pro-
file because of transaction costs) and no dramatic increase in seigniorage to
cushion the effect of the fall in q. Seigniorage is actually negative in present
value under the explosive paths, as the public dumps its currency holdings
(row 12 of Table 4). Under explosive paths with κ ≥ 10−4 the central bank,
therefore, becomes insolvent: it can no longer stick to the Taylor rule and
at the same time honor its liabilities without an intervention from the fiscal
authority.
VIII. SELF-FULFILLING SOLVENCY CRISES
As we have already observed, a central bank cannot guarantee determi-
nacy of the price level in the absence of fiscal backing. Our detailed sce-
narios in the previous sections have all assumed (except in the κ > 0 cases)
that this backing was present. But even when the backing is present, a cen-
tral bank that is firmly committed to not accepting (or incapable of drawing
on) fiscal support, in the sense of capital injections from the treasury, can
create indeterminacy in the price level. The problem is that commitment
to a policy rule that stabilizes inflation, like a Taylor rule with large coeffi-
cient on inflation, may under certain conditions require a capital injection
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 32
from the treasury to be sustainable. If the central bank, to avoid the cap-
ital injection, switches policy so as to generate more seigniorage, multiple
non-explosive equilibria can arise. We give examples of this possibility in
this section. In our calibrated version of the U.S. economy the particular
form of multiple equilibria we consider would only arise for levels of the
central bank’s balance sheet larger than the current one, although this result
depends crucially on the properties of the demand for currency.
What if the public believes that, were the central bank to face the issue of
solvency, it would resort to seigniorage creation? Entertaining the possibil-
ity of central bank’s insolvency would then lead the public to expect higher
future inflation and nominal interest rates. These expectations would re-
sult in a lower value of long term nominal assets today, so that the central
bank’s assets qBC could become worth less than its interest bearing liabil-
ities V. If the current present discounted value of seigniorage is not large
enough to cover this gap, the central bank may have to resort to raising
more seigniorage, thereby validating the initial belief. The larger is the size
of the central bank’s balance sheet, and the longer its duration, the larger
is the gap in qBC −V that would arise because of future expected inflation,
and the likelihood of these alternative equilibria.
If there is the possibility of indeterminacy, these multiple equilibria can
take many forms. We focus on a particular type of multiple equilibria,
where agents expect that at time t = T the central bank will change its infla-
tion target to π for a period ∆, and revert to the old rule with inflation target
π afterwards (for t > T + ∆). The appropriately modified version of equa-
tion (14) provides the solution for the future path of interest rates. Given
the path for rt we can solve for all other endogenous variables exactly as in
the model above. We can in particular obtain, under this alternative equilib-
rium, the value of long term assets q0(π), the initial price level P0(π), and
the present discounted value of seigniorage in real terms at time 0, which
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 33
we can call PDVS0(π) =∫ ∞
0(
Mt
Mt+ n)
Mt
Pte−∫ t
0 (ρs−n)dsdt. All of these ob-
jects will be a function of π (and of T and ∆ as well). For each T and ∆,
we can then find what expected future inflation π needs to be to generate a
solvency crisis, that is, we can look for solutions of
q0(π)BC0− −V0
P0(π)+ PDVS0(π) = 0, (40)
if any exist. These solutions are possible self-fulfilling solvency crises, in
the sense that the expectation that the central bank will switch to a new rule
with target π will produce a gap in the value of central bank’s assets minus
liabilitiesq0(π)BC
0− −V0
P0(π)that will have to be filled with future seigniorage
PDVS0(π). In order to generate this future seigniorage the central bank
will have to validate the public expectations and switch temporarily to the
rule with higher inflation target.
Using our simple calibrated model we search for these alternative equi-
libria. In particular, for given π, T, and ∆ we find the minimum level of the
balance sheet BC for which equation (40) has a solution. The top panel left
panel of figure 8 shows this minimum balance sheet level (relative to the
current level) as a function of inflation in there alternative regime (π ), and
the duration of the alternative regime (∆). To repeat, these are the balance
sheet levels for which the solvency constraint would become binding if the
public expects a regime with inflation π and duration ∆ to materialize in
T = 3 years.
The figure shows that under the baseline money demand calibration,
these threshold balance sheet limits are much larger than the current one.
The left middle and bottom panels of figure 8 explain why this is the case.
The middle panel shows what happens toq0(π)BC
0− −V0
P0(π)in the alterna-
tive equilibrium under the current balance sheet size. The figure shows that
for large enough π and duration ∆ the real value of assets minus interest
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 34
bearing liabilities does become significantly negative. The bottom panel
of figure 8 show that for these values the level of seigniorage PDVS0(π)
overshadows this balance sheet loss, however. Hence the results in the top
panel: the size of the balance sheet would have to be much larger than the
current one for the balance sheet loss to be of the same size of the increase in
seigniorage. In other words, under the current level of the balance sheet, the
type of alternative equilibria we consider cannot arise because the increase
in seigniorage triggered by the temporary higher inflation regime is larger
than the balance sheet loss caused by the fall in q.18 The Laffer curve in the
left panel of figure 7 shows why the increase in seigniorage is so large un-
der the baseline parameterization of money demand: even for large interest
rates money demand asymptotes to positive values, and seigniorage grows
linearly with inflation.19
Under the alternative money demand the threshold balance sheet lim-
its are much closer to one, as a consequence of the fact that the increase in
seigniorage is smaller. The right panel of figure 7 shows that seigniorage
peaks at about 40 percent steady state inflation, and becomes zero for in-
flation larger than one hundred percent. Hence seigniorage increases much
less with π than in the baseline case, as shown in the bottom right panel of
figure 8 (since the duration of the alternative regime is relatively short we
find that seigniorage still increases for values of π up to 500 percent). Still,
we find that the threshold balance sheet levels that would generate multi-
plicity are all larger than one for the (π, ∆π) values shown here. Therefore,
even for the alternative money demand multiple equilibria under the cur-
rent size of the balance sheet could be possible only if the public expected
very large inflation and/or very long duration of the alternative regime. In-
deed, we find that that for instance π = 5, that is, 500 percent inflation, for
18We searched for alternative values of T as well, and the results are not very different.19One caveat to these simulations is that we maintain the hypothesis of passive fiscal
policy even under high inflation rates, which may not be realistic.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 35
158 periods would be an equilibrium under the current level of the balance
sheet. These results are in line with what shown in the previous sections,
namely, one has to assume fairly radical scenarios for solvency to become
an issue.
IX. CONCLUSIONS
To Be Written
REFERENCES
BARRO, R. J., AND X. SALA-I MARTIN (2004): Economic Growth. The MIT
Press, second edn.
BASSETTO, M., AND T. MESSER (2013): “Fiscal Consequences of Paying In-
terest on Reserves,” Federal Reserve Bank of Chicago working paper No. 2013-
04.
BENHABIB, J., S. SCHMITT-GROHÉ, AND M. URIBE (2001): “The Perils of
Taylor Rules,” Journal of Economic Theory, 96, 40–69.
CARPENTER, S. B., J. E. IHRIG, E. C. KLEE, D. W. QUINN, AND A. H.
BOOTE (2013): “The Federal Reserve’s Balance Sheet and Earnings: A
primer and projections,” FEDS Working Paper, 2013-01.
CHRISTENSEN, J. H. E., J. A. LOPEZ, AND G. D. RUDEBUSCH (2013): “A
Probability-Based Stress Test of Federal Reserve Assets and Income,” Fed-
eral Reserve Bank of San Francisco Working Paper 2013-38.
COCHRANE, J. H. (2011): “Determinacy and Identification with Taylor
Rules,” Journal of Political Economy, 119(3), 565–615.
CORSETTI, G., AND L. DEDOLA (2012): “The mystery of the printing press:
self-fulfilling debt crises and monetary sovereignty,” mimeo Oxford Uni-
versity.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 36
DEL NEGRO, M., F. SCHORFHEIDE, F. SMETS, AND R. WOUTERS (2007):
“On the Fit of New Keynesian Models,” Journal of Business and Economic
Statistics, 25(2), 123 – 162.
FEDERAL RESERVE SYSTEM (2014): “The Federal Reserve Banks. Com-
bined Financial Statements as of and for the Years Ended De-
cember 31, 2013 and 2012 and Independent Auditors’ Report,”
http://www.federalreserve.gov/monetarypolicy/files/BSTcombinedfinstmt2013.pdf.
HALL, R. E., AND R. REIS (2013): “Maintaining Central-Bank Solvency un-
der New-Style Central Banking,” manuscript.
ORPHANIDES, A. (2002): “Monetary-Policy Rules and the Great Inflation,”
The American Economic Review, 92(2), 115–120.
SIMS, C. A. (2005): “Limits to Inflation Targeting,” in The Inflation-Targeting
Debate, ed. by B. S. Bernanke, and M. Woodford, vol. 32, chap. 7, pp. 283–
310. NBER Studies in Business Cycles.
(forthcoming): “Paper Money,” American Economeic Review, 103(2),
1–24.
WOODFORD, M. (2001): “Fiscal Requirements for Price Stability,” Journal of
Money, Credit and Banking,, 33(1), 669–728.
APPENDIX A. RULE FOR REMITTANCES
The central bank is assumed to follow a rule for remittances, which em-
bodies two principles: i) remittances cannot be negative, ii) whenever posi-
tive, remittances are such that the central bank capital measured at historical
costs remains constant in nominal terms over time, that is:20
K =(
qBC −V −M)
ent = constant. (41)
The historical price q evolves according to
˙q = (q− q)max{
0,BC
BC + δ + n}
(42)
20Hall and Reis (2013) use a similar rule, but measure capital at market prices.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 37
where the max operator is there because q changes only if the central bank
is acquiring assets (recall that bonds depreciate at a rate δ and that BC is
defined in per capita terms, so that BC = −(δ+ n)BC implies that the central
bank is letting its assets mature). Differentiating condition (41) above and
using the central bank’s budget constraint (24), one obtains a condition for
nominal remittances:
PτC = (χ− δ(q− 1)) BC +
(˙q− (q− q)
(BC
BC + δ + n))
BC − rV. (43)
This condition resembles closely the accounting practice of central banks.
The first term, (χ− δ(q− 1)) BC, measures coupon income χ net of the
amortization of historical costs δ(q − 1), times the par value of bonds BC.
The second term equals the realized gains/losses,−(q− q)(
BC
BC + (δ + n))
BC,
from assets sales (that is,BC
BC ≤ −(δ + n)), since in this case ˙q = 0. When
the central bank is acquiring assets (BC
BC > −(δ + n)) this second term is
zero because ˙q and (q − q)(
BC
BC + δ + n)
cancel each other out. The last
term, rV, captures the interest paid on reserves. Whenever net income (the
left hand side of equation 43) is negative, the central bank would have to
extract negative remittances from the fiscal authority to keep its capital con-
stant. If it cannot do this, its capital declines. Whenever internal accounting
rules prevent capital from declining a deferred asset is created. Remittances
remain at zero until this deferred asset is extinguished (i.e., capital is back
at the original level). Hence our rule for remittances is
τCt = max
{0, (χ− δ(q− 1))
BC
P
+
(˙q− (q− q)
(BC
BC + (δ + n))
BC
P
)− r
VP
}I{K≥K0}, (44)
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 38
where I{K≥K0} is an indicator function equal to one only if current capital K
is at least as large as initial capital K0 (that is, the deferred asset has been ex-
tinguished). In practice central bank’s capital will not be constant over time,
but will likely grow along with nominal income. This implies a net influx
of resources for the central bank. At the same time a fraction of net income
is devoted to pay dividends on this capital.21 Moreover the central bank
also has operating expenses. We ignore these issues in computing the sim-
ulated path of remittances since it would further complicate the description
of the remittance rule, and also quantitatively they are not very important
in terms of the simulated path for remittances.
In order to compute the path for remittances implied by expression (44)
we need to compute paths forBC
Pand
VP
. For the former, we make assump-
tions about the path of the central bank’s assets. We make two assumptions
about the future path of BC:
BC
P=
−(δ + n)
BC
P(1)
−(δ + n + s)BC
P(2)
, for t ≤ T, (45)
where T is the time when the size of reserves has reached the early 2008 level
(adjusted for inflation, population and productivity growth), after whichBC
Pgrows with productivity (i.e.,
BC
Pe−γt is constant over time, yielding
BC
P= (γ +
PP)
BC
P). Under assumption (1) the central bank lets its holdings
of government debt mature, while under assumption (2) it sells its assets at
a rate s per year (we set s = .2). Neither assumption is realistic in the case
of the U.S. (BC has increased in 2013) but the point is to show that different
21In the U.S. the central bank’s capital is a fixed fraction of the capital of the mem-
ber banks, and dividends are 6% of capital (see Carpenter, Ihrig, Klee, Quinn, and Boote
(2013)). Note also that according to our notation dividends are included in τC, this quantity
being the total amount of resources leaving the central bank in any given period. In this
sense referring to τC as “remittances” to the fiscal authority is not entirely appropriate.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 39
future paths for sales can yield quite different paths for τC in the short run,
even though the present value of resources remitted to the fiscal authority
τC is the same. Given remittances and the path forBC
Pwe use the budget
constraint (24) to compute the evolution of reserves in real terms:
.(VP
)= (r− n− P
P)
VP
− (χ + δ− (n + δ)q)BC
P+ q
BC
P−(
n +MM
)MP
+ τC. (46)
A legitimate question (also posed by Hall and Reis (2013)) is whether
a rule like (44) keeps the central bank’s capital measured at market prices,
namely
K =(
qBC −V −M)
ent. (47)
stationary. Using the budget constraint (24) we write the evolution of de-
trended capital in real terms (KP
e−(γ+n)t) as
d
(KP
−(γ+n)t)
= (ρ− n− γ)
(KP
e−(γ+n)t)+
(r
MP− τC
)e−γt (48)
So the rule that stabilizesKP
e−(γ+n)t is
τC = rMP
+ (ρ− n− γ)
(KP
e−(γ+n)t)
= r(
qBP− V
P
)−(
PP+ n + γ
)(KP
e−(γ+n)t)
. (49)
This rule has quite different implications for remittances relative to the rule
(44) outside of steady state, but at steady state the two coincide. In fact, at
steady state q = q = 1, K = K, and χ = r, hence the term r(
qBP− V
P
)coincides with the right hand side of expression (44). The remaining term
(π + n + γ)
(KP
e−(γ+n)t)
accounts for the fact that capital increases with
inflation, productivity, and population growth (as discussed above), a factor
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 40
which we ignore in (44). If we did properly account for it, expressions (49)
and (44) would be consistent with each other at least at steady state.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 41
TABLE 1. Change in steady state after 2% inflation scare
r v C M/P log(P) log(M) x dpvs
initial 0.015 12.247 1.014 0.083 0.000 0.000 0.000 0.000
new 0.075 27.386 1.012 0.037 0.080 -0.726 0.020 2.667
duration 2.5 5 10 20
A0 = 3 A loss -0.14 -0.28 -0.52 -0.92
gap -0.22 0.35 1.13 1.98
duration 2.5 5 10 20
A0 = 6 A loss -0.14 -0.28 -0.52 -0.92
gap 0.57 1.71 3.25 4.95
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 42
TABLE 2. Parameters
normalization, foreign assets
Y− G = 1 F0 = 0
initial assets, reserves, and currencyBC
P= 0.327
VP
= .207MP
= .104
discount rate, reversion to st.st., population and productivity growth
β = 0.01 γ = 0.01
ϕ1 = 0.750 n = 0.0075
monetary policy
θπ = 2 θr = 1
π = 0.02
money demand
ψ0 = 2.31*10−6 ψ1 = -0.055bonds: duration and coupon
δ−1 = 6 χ = 0.035
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 43
TABLE 3. Central bank’s resources under different simulations
(1) (2) (3) (4) (5) (6) (7)
qB/P−V/P
PDVseigniorage (1)+(2)
τC
(averageremittance)
q ∆M/P B/B
Baseline calibration
(1) Baseline scenario 0.146 0.998 1.144 0.0026 1.08
(2) Higher rates (β) 0.133 0.169 0.302 0.0033 1.06 -0.006 14.42
(3) Higher rates (γ) 0.142 1.287 1.429 0.0031 1.06 0.003 64.59
(4) Higher rates (β)/same intercept 0.097 0.241 0.338 0.0037 1.01 -0.025 5.75
(5) Higher rates (γ)/same intercept 0.105 1.573 1.677 0.0037 1.01 -0.017 24.56
(6) Inflation scare 0.068 1.007 1.075 0.0024 0.92 -0.024 7.81
(7) Explosive path 3.124 Inf Inf Inf 0.88 3.043 Inf
Higher θπ
(8) Higher rates (β)/same intercept 0.115 0.216 0.331 0.0035 1.04 -0.019 9.93
(9) Inflation scare 0.085 0.992 1.077 0.0024 0.96 -0.021 10.10
(10) Explosive path 6.124 Inf Inf Inf 0.64 6.120 Inf
Lower θπ
(11) Higher rates (β)/same intercept -0.055 1.162 1.107 0.0132 0.52 -0.024 3.38
(12) Inflation scare -0.012 1.038 1.026 0.0025 0.67 -0.021 3.90
(13) Explosive path 0.241 Inf Inf Inf 1.05 0.106 Inf
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 44
TABLE 4. Central bank’s resources under different simula-
tions. Calibration with money demand = 0 for r > 100%
(1) (2) (3) (4) (5) (6) (7)
qB/P−V/P
PDVseigniorage (1)+(2)
τC
(averageremittance)
q ∆M/P B/B
Baseline calibration
(1) Baseline scenario 0.146 0.214 0.360 0.0008 1.08
(2) Higher rates (β)/same intercept 0.060 0.033 0.093 0.0010 1.01 -0.062 2.31
(3) Inflation scare 0.012 0.215 0.227 0.0005 0.92 -0.080 2.44
(4) Explosive path 0.079 -0.094 -0.015 -0.0000 0.88 -0.002 0.92
Higher θπ
(5) Higher rates (β)/same intercept 0.085 0.017 0.103 0.0011 1.04 -0.048 3.77
(6) Inflation scare 0.029 0.121 0.150 0.0003 0.96 -0.077 2.27
(7) Explosive path 0.001 -0.097 -0.096 -0.0002 0.64 -0.002 0.78
Lower θπ
(8) Higher rates (β)/same intercept -0.111 0.095 -0.016 -0.0002 0.52 -0.080 0.97
(9) Inflation scare -0.072 0.299 0.227 0.0005 0.67 -0.081 1.64
(10) Explosive path 0.133 0.041 0.175 0.0004 1.05 -0.002 6.46
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 45
FIGURE 1. A scatter plot of short term interest rates and M/PC
0 0.05 0.1 0.15 0.20.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
M/P
C
r
2013
2008
Modelpre 1980 datapost 1980 data
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 46
FIGURE 2. Short term interest rates and inflation: baseline vs
higher rates
Nominal Real
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 20230
1
2
3
4
5
6
r
BaselineHigher ratesHigher rates (same intercept)
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 20230
0.5
1
1.5
2
2.5
3
ρ
BaselineHigher rates
Inflation
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023−0.5
0
0.5
1
1.5
2
2.5
3
π
BaselineHigher ratesHigher rates (same intercept)
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 47
FIGURE 3. seigniorage and M/PC
Data Model
0.05
0.055
0.06
0.065
0.07
0.075
0.08
0.085
0.09
0.095
0.1
M/PCE −− left axis
1980 1985 1990 1995 2000 2005 20100
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
S/Y −− right axis
0.06
0.065
0.07
0.075
0.08
0.085
0.09
M/PC −− left axis
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023−8
−6
−4
−2
0
2
4x 10
−3
Seigniorage −− right axis
Notes: Paths for seigniorage and real money balances in right hand panel are obtained underbaseline simulation.
FIGURE 4. Paths for remittances
Baseline Inflation Scare
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 20232
3
4
5
6
7
8
9
10x 10
−3
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 20230
0.5
1
1.5
2
2.5
3
3.5x 10
−3
Notes: Each panel shows remittances under two assumptions for the path of assets BC: under thefirst assumption (solid line) the central bank lets its assets depreciate, while in the second one itactively sells assets at a rate of 20 percent per year. The paths in the left and right panels areobtained under the baseline and “inflation scare” scenario, respectively.
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 48
FIGURE 5. The effect of different inflation responses in inter-
est rate rule under different scenarios
Higher β/same interceptShort term interest rates
Inflation scareShort term interest rates
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 20230
5
10
15
20
25
r
Baselineθπ=2
θπ=3
θπ=1.05
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 20230
2
4
6
8
10
12
14
16
18
r
Baselineθπ=2
θπ=3
θπ=1.05
Inflation scareInflation
Explosive pathsShort term interest rates
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023−2
0
2
4
6
8
10
12
14
π
Baselineθπ=2
θπ=3
θπ=1.05
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 20230
5
10
15
20
25
r
Baselineθπ=2
θπ=3
θπ=1.05
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 49
FIGURE 6. Case where money demand = 0 for r > 100%
Short term interest rates and M/PC Seigniorage and real money balances
0 0.05 0.1 0.15 0.20
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
M/P
C
r
2012
2008
Modelpre 1980 datapost 1980 data
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
M/PC −− left axis
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023−0.025
−0.02
−0.015
−0.01
−0.005
0
0.005
Seigniorage −− right axis
Notes: Paths for seigniorage and real money balances in right hand panel are obtained underbaseline simulation.
FIGURE 7. Laffer curves
Baseline Calibration withmoney demand = 0 for r > 100%
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20
0.02
0.04
0.06
0.08
0.1
0.12
Se
ign
iora
ge
piss
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
0
0.2
0.4
0.6
0.8
1
1.2x 10
−3
Sei
gnio
rage
piss
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 50
FIGURE 8. Self-fulfilling solvency crises
Baseline Calibration withmoney demand = 0 for r > 100%
Threshold Balance Sheet Limit% of current B
qB−V
seigniorage
Notes: The figure shows 1) Top panel: the level of the balance sheet (relative to the current level) forwhich multiple equilibria are possible; 2) Middle panel: the level of qB−V as a fraction of incomefor the current balance sheet size under alternative scenarios; 3) Bottom panel: the level ofseigniorage as a fraction of income under alternative scenarios; as a function of a) inflation in therealternative regime (π), b) the duration of the alternative regime (∆). In all simulation the alternativeregime is expected to start after 3 years. The left and right figures are for the baseline and alternativetransaction technology, respectively. .
CENTRAL BANK’S BALANCE SHEET. PRELIMINARY DRAFT 51
E-mail address: [email protected]
E-mail address: [email protected]